Partnership for Education? A Case Study on Tumaco, Colombia on a North-South Co-operation for Education in Conflict Areas

(1)

Partnership for Education?

A Case Study on Tumaco, Colombia on a North-South Co-operation for Education in

Conflict Areas

Tobias Immanuël Alexander Wilbers

Master’s Thesis in Latin American Studies

Department of Literature, Area Studies and European Languages

Faculty of Humanities

Supervisor: Jemima Garcia-Godos UNIVERSITY OF OSLO

May 2015

(2)

II

(3)

III

Partnership for Education?

A Case Study on Tumaco, Colombia on a North-South Co-operation for Education in

Conflict Areas

By: Tobias Immanuël Alexander Wilbers

Latin American Area Studies

Department of Literature, Area Studies and European Languages

Faculty of Humanities

Supervisor: Jemima Garcia-Godos

UNIVERSITY OF OSLO Turned in May 15th.

2015

(4)

IV

Partnership for Education: A Case Study on Tumaco, Colombia on a North-South Co- operation for Education in Conflict Areas

Tobias Immanuël Alexander Wilbers http://www.duo.uio.no/

Print: Reprosentralen, University of Oslo

(5)

V

Abstract

This thesis contributes to the subject of north-south partnerships between non-governmental development organisations (NGOs) working with education in areas affected by armed conflict and humanitarian challenges. The increased numbers of NGOs and other international organisations such as the UN working with development is used as a demand for better co- ordination of development projects, both to avoid project overlapping, but also to increase the impacts of the projects. The case study focuses on an ongoing partnership in Tumaco, located on the Pacific coast in South-Western Colombia, which is experiencing high levels of armed conflict, leading to high numbers of forced displacements. The studied partnership consists of two northern development NGOs, as well as two local NGOs and a Community-Based Organisation (CBO), but also includes co-operation with Colombian education authorities, local Education Institutions and universities. This demonstrates a complex dynamic between the partners, and poses challenges to issues such as balance in the co-operation in terms of decision-making and project management.

The partnership’s aim is to enrol internal displaced persons and marginalised population into the public school system with the use of flexible education models and institution strengthening of the education institutions. The study’s central subject is to explore to which degree the partnership is complying with its prerequisites. Based on the case study, the paper concludes that the partnership is a necessary mechanism to increase the desired outcomes of the projects. The partners have different levels of participation in the co-operation. The study shows that the studied northern NGOs have a well-established relationship with the local community organisations, while the relations with the education authorities is a more formal, and is perceived as giving public legitimacy to the project. The study contributes to a discussion on how development actors co-operate on different levels to assist internal displaced and vulnerable persons in areas with high levels of security challenges with education.

(6)

VI

Resumen

Esta tesis es una contribución al tema sobre la colaboración entre organizaciones no- gubernamentales (ONG) del Sur y el Norte enfocadas en desarrollo, que trabajan con educación en áreas afectadas por conflicto armado y desafíos humanitarios. La incrementada cifra de ONG y otras organizaciones internacionales como la ONU que trabajan en desarrollo ha sido usada como una demanda para mejorar la coordinación de proyectos de desarrollo.

Esta demanda está basada en una teoría de que es necesario evitar que los proyectos se traslapen, y también para aumentar el impacto de los proyectos. El estudio de caso trata una alianza que origina en Tumaco, situada en la costa pacífica colombiana, que experimenta altos niveles de conflicto armado, algo que conlleva a frecuentes desplazamientos forzados internos de personas. La estudiada asociación consiste de dos ONG del norte, dos ONG colombianas y una corporación de grupos comunitarios. Además la asociación incluye a autoridades de educación, instituciones educativas y universidades. El estudio da prueba de una dinámica compleja entre los socios, y plantea desafíos relacionados a temas como equilibrio en la toma de decisiones y gestión del proyecto educativo.

El enfoque de la asociación es incluir niños, niñas, jóvenes y adolescentes desescolarizados al sistema educativo, además de trabajar con la calidad de la educación existente. Esto se hace a través del uso de modelos de educación flexible y el fortalecimiento de las instituciones educativas. El tema central del estudio es explorar hasta qué niveles la asociación cumple con sus prerrequisitos. La conclusión del estudio de caso demuestra que la asociación es un mecanismo necesario para aumentar los resultados deseados del proyecto. Los socios tienen diferentes niveles de participación en la alianza. El estudio demuestra que las ONG del norte tienen una relación bien establecida con las organizaciones de la comunidad, mientras las relaciones con las autoridades educativas tienen un carácter más formal. Basado en lo anterior es percibido que esta relación tiene fines de dar legitimación estatal al proyecto. El estudio contribuye a una discusión sobre cómo actores de desarrollo coopera con educación a diferentes niveles para asistir desplazados internos y personas vulnerables en áreas afectadas por altos niveles de problemas con la seguridad.

(7)

VII

Preface

Lo primero que un niño traumatizado tiene que hacer es sentirse alegre, recuperar la alegría.

Translated into English, the above states that “the first thing a traumatised child should do is to recover happiness.” This is one of several statements presented to me during field work.

This serves as an example of the challenges internal displaced children in Colombia are facing, meaning that children affected by the armed conflict not always are capable of following a regular school class. Instead, they need special attention and sometimes follow flexible education models as a way of returning to school. This paper is dedicated to all the persons working to assure access to education as well as education quality in Afro-Colombian and indigenous communities in the Pacific Colombian coast; an area heavily affected by armed conflict and marginalised conditions. Some of these people gave their lives while working to improve living conditions for their children, families, friends and neighbours.

Keep up the good work RECOMPAS.

It has been a long road to get to this point of turning in the paper. I am very thankful for the support from my supervisor at the University of Oslo, Jemima Garcia-Godos, for guiding me on the way. Your professional backing and knowledge on Colombia has been indispensable for this work. I would also like to thank my supervisor in Colombia, Edgar Ramirez Monsalve from the National University of Colombia in Medellin, for initial guidance back in 2013. You introduced me to central aspects of human development and citizenship in Colombia, in which the role of education is central. My wife Adriana also deserves my sincere gratitude. In addition of giving me emotional support, you have spent hours and hours helping me with the transcriptions of the interviews. I think that there are few wives who bring their husband food while he is working at the university. I would also like to thank my brother Christian for helpful corrections of the language. Also my family has shown me emotional support – thank you! The remaining errors are solely my responsible.

Finally, I am very thankful of the help that I received from the Norwegian Refugee Council and Save the Children while staying in Tumaco. You did not only open the doors to your office, but also gave me the opportunity to share your work with education with me.

(8)

VIII

(9)

IX

List of Abbreviations

CBO – Community-Based Organisation CIA – The Central Intelligence Agency (US) CSO – Civil Society Organisation

DAC - The Development Assistance Committee

DANE - Colombian National Administrive Department of Statistics

ECHO - The European Commission's Humanitarian aid and Civil Protection department EE – Ethno-education – etno educación

EFA – Education For All

EHS – Ethno High School – Etno Bachillerato EI – Educational Institution

ES – Education Secretariat

FEM – Flexible Education Model HDI – Human Development Index

HFP – High School for Peace – Bachillerato Pacicultor IASC – Inter-Agency Standing Committee

IDB – The Inter-American Development Bank IDP – Internal Displaced Person

INEE – Inter-Agency Network for Education in Emergencies

INGDO – International Non-Governmental Development Organisation IOM - The International Organisation for Migration

(14)

XIV

MDGs – The United Nations Millennium Development Goals MoE - Ministry of Education or National Education Secretariat NGDO – Non-Governmental Development Organisation NGO – Non-Governmental Organisation

NMFA - The Norwegian Ministry of Foreign Affairs NNGO – Northern Non-Governmental Organisation

Norad – The Norwegian Agency for Development Cooperation NRC – The Norwegian Refugee Council

OCHA - The United Nations Office for the Co-ordination of Humanitarian Affairs OECD - Organisation for Economic Co-operation and Development

PAC – Proyecto Aprendiendo Crecemos – As we learn we grow project

RECOMPAS – Corporación Red de Consejos Comunitarios del Pacifico Sur - the corporation the Network of Community Councils of the South Pacific

SC – Save the Children

SCC – Save the Children Canada SCI – Save the Children International

SCiC – Save the Children in Colombia SET – Education Secretariat of Tumaco

SNGO – Southern Non-Governmental Development Organisation Udenar – The University of Nariño

UNAD – The National University Open and at Distance

UNESCO – The United Nations Educational, Scientific and Cultural Organisation

(15)

XV UNHCR – The United Nations High Commissioner for Refugees

Unicauca – The University of Cauca

UNICEF – The United Nations Children's Fund WB – The World Bank

WFP – The United Nations World Food Programme UN – United Nations

UNDP – United Nations Development Programme

VLE – Proyecto Vive la Educación – the Education Lives project

(16)

(17)

1

1 INTRODUCTION

Education is the most powerful weapon which you can use to change the world

(Nelson Mandela)

1.1 Development in the South

In a developmental perspective, what role can education play in improving the life conditions for marginalised groups affected by armed conflict, natural disasters and general poverty?

Which entity is best qualified for the organisation and implementation of education for development in the geographical “South”, and how should education be funded? What right does an international non-governmental development organisation (INGDO) have to intervene on foreign territory? How is education prioritised within developmental aid and humanitarian assistance? When it comes to education for development, is an NGO better off working alone, in partnership with other NGOs, and/or with other public and private entities?

This paper is a contribution to an ongoing discussion concerning how education for development should be managed in the South.

Education provision is considered one of several strategies for development in the South. This paper will discuss why education is considered to be important in both a humanitarian and development context, and how it is considered significant as a factor for human development.

We will present that education is a central strategy and one of the main approaches for humanitarian assistance in situations of armed conflict and natural disasters, as well as in a development context. We also show that this subject has gained increased attention from various stakeholders in the field, ranging from NGOs and governments to donors and multinational organisations working with humanitarian and development assistance.

The last couple of decades have seen an increase in the number and size of various types of non-governmental organisations (Chant & McIlwaine, 2009, p. 298), (Potter, Binns, Elliott, &

Smith, 2008, pp. 316-319) and (Fowler, 1991, p. 5). A huge number of different NGOs are currently working in the South, and it is quite common that several NGOs, aside from other UN bodies, operate in the same area within the same aid sectors, such as education. Although NGOs and UN bodies work shoulder to shoulder, several examples show that they do not co-

(18)

2

operate nor share information, neither amongst each other nor with the national government or even engage local human resources like Civil Society Organisations (CSOs) and communitarian councils (CBOs). Apart from that, NGOs attempt to demonstrate that their projects obtain the planned results. In some cases this has led to critique from the donors funding the aid projects. In some cases, we can also argue that aid organisations are better of working alone. This paper will have focus on the potential of partnership and co-operation between NGOs working with education, and between the state and local communitarian councils.

Currently there exist various incentives for effective planning of partnerships between aid agents, both including the state and the civil society. We will debate that a partnership can have several advantages, and even be indispensable for obtaining better results in aid projects, due to the sharing and interchange of knowledge, technical capacity and specialisation. The application of partnership is sometimes a prerequisite for the financing of aid projects. On the other hand, we will also discuss whether there are possible negative effects of the partnership strategy.

Armed conflict leads to forced human displacement. A common nominator for the conflicts of today is that they are internalised within the national boarders of countries, instead of being fought between nations: “Current conflicts tend to originate as localized, intrastate conflicts.

With a few notable exceptions, conflicts are now fought primarily between forces within a state” (Desai & Potter, 2006, p. 72). According to the United Nations High Commissioner for Refugees (UNHCR), in 2013 more than 50 million people were registered as displaced (2014). This is the first time since WW2 that we reach this number. In the category of displaced people we find both refugees (16,7 million people who have fled their home country), internally displaced persons (IDPs, 33,7 million of them, who have been forced to move to other areas within their country), and asylum seekers (1,2 million seeking protection in a foreign country).

1.2 Education for Development

Education provision in areas experiencing high levels of armed conflict and displacement is severely affected by the humanitarian situation, and a wide range of international and national development organisations are actively working to improve the access to and quality of

(19)

3 education. This paper examines partner organisations that co-operate to improve the access to basic education for marginalised children and youth¹ in South-western Colombia. These include non-governmental partners such as NGOs and CBOs and public education authorities and institutions. The case study illustrates that there are several factors which combined provoke a huge demand for the distribution and improvement of basic education to displaced people and a marginalised minority population.

1.2.1 Why is the Subject Important?

Today, the United Nations’ (UN) Millennium Development Goals (MDGs) have been adopted by many countries. Goal number two states: “Achieve Universal Primary Education”

(UN, 2014b). Considering the high numbers of NGOs, Tvedt states that, due to their increased international activities and presence within the development scheme, NGOs are now popular and important subjects for research activity (Tvedt, 1998).

In recent years, organisations like the Norwegian Ministry of Foreign Affairs (NMFA) and several NGOs highlight the importance of education as an approach within humanitarian response as well as development assistance. Many development organisations believe that it is necessary to co-ordinate their work. For instance, NMFA advocates the partnership strategy, both in terms of co-operating with other donors and aid receiving governments and with civil society, where working with education is considered a part of a development strategy (NMFA, 2003). Amongst others, the Norwegian Refugee Council (NRC) has already for several years categorised education as the fourth pillar within humanitarian response, together with nourishment, shelter and health services because refugee and displaced children are vulnerable and many of them need psychological stability (Crisp, Talbot, & Cipollone, 2001).

The United Nations Office for the Coordination of Humanitarian Affairs (OCHA) was established in 1991 to co-ordinate the development sector. Its main goal is to coordinate humanitarian response in emergencies, where their first pillar is to broaden the coalition for multilateral humanitarian action, advocating for partnerships (OCHA, 2014a). Another important co-ordinating entity is the Development Assistance Committee (DAC) of the Organisation for Economic Co-operation and Development (OECD). Since 1961, the OECD works to become an international forum the world´s leading aid donors, comprising of 29

1 With youth we mean persons up to the age of 26

(20)

4

members, including the European Union and countries that are identified as the largest aid contributors (OECD, 2014). Other initiatives that seek to raise co-ordination are the Paris declaration on aid effectiveness from 2005 and the Monterrey consensus from 2002. The latter is an agreement where “developing” countries that receive aid are made responsible for their own development while receiving (financial) support from “developed countries”. The Paris declaration functions as a statement for resolving partnership commitments amongst more than 60 bilateral, multilateral and civil society aid donors (Brannelly, Ndaruhutse, &

Rigaud, 2009). The large degree of different levels of co-operation and co-ordination gives incentives for nuanced research on the field.

In other words, this paper contributes to the ongoing debate on education for development, and discussed how states, national and international aid organisations can work together with the local society towards the aim of sustainable development in the South by means of formal education. It also debates the comprehension of North-South cooperation and how this can be elaborated in a result-oriented way. Further on, the paper adds to the debate on the effects of developmental assistance, posing the question whether it should be the donors, the state, parents, or the civil society that should be responsible for education planning? Or is it a combination of the former mentioned stakeholders?

1.3 Confining the Case

1.3.1 Research Topic

Based on the presented context, this paper focuses on the ability and understanding of partnership between aid agents working in the field of education. In our case, this means co- operation and co-ordination between southern and northern Non-Governmental Development Organisations (NGDOs), governments and Civil Society Organisations (CSOs). The paper elaborates on relationship dynamics between different actors in the field of delivering education in complex humanitarian situations with several risk factors.

1.3.2 Hypothesis Setting the Groundwork for the Research

In this section we mention the most central hypothesis which lays the groundwork for our research questions. By hypothesis we mean theories that help us to structure our problems of

(21)

5 discussion, and our research project in general. The hypothesis also serve as foundations for the design of the further investigation (Johannessen, Tufte, & Christoffersen, 2010, p. 86).

 NGOs work with education where the national government has low levels of responsibility. The NGO implementation for education is not sustainable, due to changing priorities by donors. This leads to development projects being terminated due to insufficient funding, interfering with the idea of a long-term focus and follow-up measures necessary for a high success rate for the project.

 NGOs work within financial restrictions on how and where their education projects are to be elaborated, linked to the donor’s interests, priorities, and policies on how financial contributions to development are to be applied. This in turn may affect the relation between the donor and the receiving development organisations because the latter may have to change their strategies to comply with donor interests.

 It can be challenging to co-operate with national governments, if the labour of the NGOs is not acknowledged (Miller-Grandvaux, Welmond, & Wolf, 2002). This compromises the influence the education projects have on the local society as well as possible influence on public regulations towards the education sector.

 There is little co-operation between development organisations active in the same area and/or with the same sector, such as education. This denotes that the organisations work independently from each other, which again can lead to projects having lesser impact according to the theory that organisations can increase their impact when working together and sharing information and strategies.

 The idea of partnership as a condition for receiving development funds, is not actually working. By this we mean that northern and southern NGOs do not comply with a reciprocal balanced influence on the planning of development projects. Instead, northern NGOs and donors dominate the partnership since they are financing the education projects.

1.3.3 The Research Questions

Based on the previous hypothesis, the following research questions have been elaborated and will guide the data collection:

(22)

6

1. How do the partners perceive the North-South co-operation for education and the idea of partnership?

2. Why do the partners co-operate for education, and what are the foundations for co- operation?

3. How are the roles defined in terms of co-ordination?

It is not possible to investigate everything concerning a partnership; therefore we have to develop some indicators that will help us in the search for answers to our research questions.

Hence, we have constructed the following indicators for the study: Understanding of partnership; Equality in decision making/Transparency; Communication; Local empowerment; Co-ordination; Expectations/common goals; Challenges; and Sustainability.

By sustainability we mean that the partnership is well-functioning and will continue to exist.

The indicators are based on a theoretical framework used to identify partnership strategies amongst development organisations, presented in chapter 3. In chapter 2 we further present how the indicators are applied to the study.

Topics for the study and operationalisation of the research questions

The above mentioned indicators are similar to the topics of the study, which helps to structure the data collection:

 Training of local entities such as CSOs, CBOs and Education Institutions (EIs), implementing local empowerment.

 The idea of partnership and balanced participation.

 Advocacy towards the education system.

 Relations between the partners.

 Sustainability – (common) goals for the future, the role of education authorities.

 The amount of co-operation between the partners and division of responses.

 Agendas and outcomes each partner has with the co-operation. Do they have common priorities?

 Financing and the role of the donor.

 How differences between humanitarian assistance and development aid influences the strategies of the involved NNGOs, and the amount/existence of cultural differences between all involved partners in the project.

 Level of co-operation with UN-bodies such as UNHCR, UNICEF and UNDP

 Education quality

(23)

7

1.3.4 Justification of Research Site

The municipality of Tumaco is located on the Colombian Pacific coast. Compared to the rest of the country, the area is severely affected by armed conflict, and has large areas with difficult access due to obstruction by water and poor infrastructure. There was more research on the Pacific coast before the armed conflict reached the area, for instance by the Colombian social anthropologist Arturo Escobar. Due to the ongoing armed conflict, there is little investigation on the area. A partnership study on Tumaco was chosen since the area has a high presence of NGOs and international development organisations. It is reported that around 20 NGOs are working in the area, mainly independent from governmental agencies and other NGOs. Simultaneously it is an area with high levels of IDPs, and large amounts of out-of-school children and youth are reported in the area. The predominantly Afro-Colombian population tends to feel itself abandoned by the rest of the country (Personal communication with ANSPE staff, 2012).

The World Bank reports that there are several countries showing great discrepancy in the case of education quality and enrolment numbers, listing several countries where the illiteracy rate amongst the total adult population is less than 50% such as in the cases of Niger (15%), Chad (37%) and the Ivory Coast (41%) (2013b). Currently, many countries are affected by different levels of armed conflict or show great deficits in education services and general development with no efficient public education institutions. This sometimes leads the NGOs themselves to take charge in tuition, a form of non-formal education.

The case of Colombia differs from the mentioned African countries, since it has strong governmental institutions and there is an estimated national literacy rate of 94% according to CIA (2014). Public school enrolment figures are high. Education is managed through the National Ministry of Education (MoE), which has secretariats throughout the country on local, regional and departmental levels. Therefore, national and international NGOs work towards the strengthening of the public education system, where they contribute with financial support and technical expertise rather than offering non-formal education. This dynamic can be compared to a Public-Private Partnership (PPP), since several INGOs and national NGOs work together with UN bodies as well as other NGOs, state entities, education institutions (EIs) and public universities, forming different levels of co-operation and partnerships within education. Nevertheless, Colombia contrasts from other states faced with armed conflict and complex emergency situations because the Colombian government has

(24)

8

developed frameworks for the protection of victims of the conflict, but has challenges in implementing these policies (NRC, 2014b).

Colombia is not a typical recipient of development aid, as it is categorised as an upper middle-income country and therefore considered capable of providing adequate education services. As of 2014, Colombia is the fourth biggest economy in Latin America and has the third biggest population in the region, with more than 46 million habitants (CIA, 2014).

According to a source in a Colombian NGO working with education, the former president Alvaro Uribe (who governed between 2002 and 2010), informed the international society that the country was in no position of needing international aid of any kind (Education manager Colombian NGO, 2013). Nor did he admit that the country was facing a critical situation, with millions of people being internally displaced.

When one takes into account the means of measurement, the case study shows that one will get different numbers for school enrolment when one changes the focus from a national to a regional or local level, if one leaves out the pupils above school-age that are enrolled in the education system.

Colombia is currently experiencing a complex humanitarian situation, leading to situations where children and youth are deprived their right to school attendance. Furthermore, the country is ranged as one of the most unequal countries in the world in terms of income distribution. For the measurement of income inequality, the GINI index is commonly used.

The higher the number, the greater is the inequality of income on a national scale (the index ranges from 0 = no equality, to 100 = maximum equality). WB figures for 2010 show that Colombia has a GINI of 55,9. As a comparison, Norway has an index of 25,79 (2014).

1.3.5 The Structure of the Thesis

This paper has the following structure:

 In chapter 2 the methodological framework is presented, introducing the technical design of the study and discusses the data collection method.

 Chapter 3 gives a theoretical background for the role of education in development aid and humanitarian assistance and presents the outlines of a theoretical framework for NGO partnerships and how this can be applied to education.

(25)

9

 Chapter 4 contextualises the current situation in Colombia in terms of armed conflict, the humanitarian situation, and how education is affected. It also discusses the role of the Colombian government within education and international co-operation for aid.

 Chapter 5 presents the partnership with its partners based on the collected data

 Chapter 6 analyses the partnership and the roles of the involved partners

 Chapter 7 is reserved for the general conclusions

Summary

This chapter positioned the case study in a theoretical framework, highlighting the increasing numbers of NGOs working with humanitarian assistance and development, including education. The large amount of development organisations is said to give support for higher levels of co-ordination amongst the NGOs and other stakeholders in development aid, creating new challenges for decision-making and the management of development projects.

(26)

10

2 METHODOLOGICAL FRAMEWORK

Introduction

This chapter centres on the design of the methodological framework, including planning, collection and analysis of data necessary for the structuring of the research project. Firstly, the selection of the case study-method is justified. Secondly, the discussion will centre on the available choices of sources for the research project and how they can be applied to the study.

Finally, we highlight some challenges centred on the data collection and the role of the investigator in the field. Therefore, we need to elaborate a research design for our study as a way of assuring reliability and validity of the project.

2.1.1 Research Design

We are studying a partnership between two northern development NGOs (NDNGOs), who work as partners with Colombian NGOs, a Colombian local CBO as well as local and national education authorities and education institutions (EIs). For this purpose, we need to elaborate a plan for the investigation. According to Johannessen et al. (2010, pp. 73-74), a research design relates to the procedure that includes all aspects encompassing the project, ranging from the planning phase to the final phase of handing in the project. Yin defines a research design as “the logic that links the data to be collected to the initial questions of study” (Yin, 2009, p. 24). Further on, it is necessary to define what and who we want to investigate, and also how the project is to be completed.

Yin (2009, p. 8) states that the selection of method depends on three conditions: (1) The type of research question posed, (2) The extent of control an investigator has over actual behavioural events, and (3) the degree of focus on contemporary as opposed to historical events. The case study approach is suitable as a manner to gather as much information as possible about a limited subject, such as an occurrence within for instance an organisation. It serves to study one or several incidents thoroughly, giving the researcher the possibility to complete a detailed study in which one identifies the most relevant factors given the research questions (Johannessen et al., 2010, pp. 75, 85). Here one is also given the possibility to complete a thorough study to identify the most relevant aspects for the researcher and his or

(27)

11 hers research questions concerning specific aspects around a given case (Ragin & Amoroso, 2011, p. 115), something that fits our partnership study.

Our case concerns temporary events, something that supports the use of case study. When choosing the case study method, the advantage is to have a range of different data collections options available, meaning that the method serves to study an event using multiple sources (Yin, 2009, p. 11).

2.1.2 Validity and Reliability

As a way of ensuring the quality of the research design and thereby the investigation in the design process of the case study, it is necessary to have some criteria for trustworthiness, credibility, confirmability and data dependability (Yin, 2009, p. 40). These criteria can be met by taking into consideration the validity and reliability of the study.

To construct validity, the concepts being studied need to be identified using correct operational matters. Also, when linking relationships, we need to assure that certain conditions really lead to other conditions, and thereby we avoid spurious associations. This means that we have to choose relevant indicators for the study that reflect the study’s questions. Furthermore, the research has to be carried out in way that gives it generalisability.

The latter refers to an analytic generalisation rather than a statistical generalization. A study has both internal and external validity, where the former refers to a causal relationship rather than a spurious one, whereas the latter deals with the extent in which it is possible to generalise from the study’s findings (Yin, 2009, pp. 40-42). Finally, validity relates to how well, or how relevant the data presenting the phenomena that is to be investigated actually is (Johannessen et al., 2010, p. 408).

For a study to be reliable, “[the researcher has] to be sure that, if a later investigator followed the same procedures as described by an earlier investigator and conducted the same case study all over again, the later investigator should arrive at the same findings and conclusions” (Yin, 2009, p. 45). This means that the research procedures have to be well documented, making it possible for a potential external reviewer to understand the different steps made in the procedure. Throughout the thesis, validity and reliability will be commented on.

(28)

12

2.2 Yin’s Five Point Research Design

According to Yin (2009, p. 27), a research design needs five components: (1) A study´s questions; (2) its propositions, if any; (3) its unit(s) of analysis; (4) the logic linking the data to the propositions; and (5) the criteria for interpreting the findings. This research design is to have a logical sequence, connecting the empirical data with the initial research questions.

To identify the data that is to be collected, it is necessary to establish a theory depicting the types of answers one expects linked to the chosen research questions (Yin, 2009, pp. 35-36).

By theory it is meant here hypothesis on possible correlations, a possible theory that has yet to be confirmed. We need to be clear of what we look for, and therefore we need hypothesis to guide the research ahead. Here we have to ascertain that we do not merely select the data that confirm that our theory (hypothesis) is correct, and by that consciously or unconsciously refrain to take into consideration the information that not suits our arguments (Kjeldstadli, 1999, p. 138). Since the subjects for the research tend to be complex, it is necessary with a theory that serves to simplify correlations. Generalisations, models and general rules can all be seen as theory, according to Nordby (2008).

2.2.1 Research Questions

The study’s questions presented in the introductory chapter set the bases for the selection of the methodological framework, and fit well in the case study approach with their classification as “how” and “why” questions.

2.2.2 The Study’s Propositions

The second component focuses on the study´s propositions, if any. According to Yin (2009, p.

28), although we have stated our research questions, we still need to figure out what we should study to be able to answer the questions, in other words, determine what helps us to structure where to look for relevant evidence. Regarding our case, a proposition is that northern and southern (development) organisations work together to strengthen the quality of and enrolment in education, leaning towards the understanding of that the entities work better together. Another proposition is that northern development organisations depend on local knowledge and recourses for a successful implementation of their development projects.

These propositions help us to narrow down the units for our case study.

(29)

13

2.2.3 Units of Analysis

The third component in the research design discusses the unit(s) of analysis. This section defines what the “case” is, something that according to Yin is a fundamental problem for many investigators to be consistent with (2009, p. 29). For our project, the case is the partnership established between several entities working with education directed to marginalised children and youth, aimed mainly towards the displaced population and minority groups. Therefore, in our case, we are dealing with an embedded single-case design which includes several subunits of analysis (Yin, 2009, p. 46). In other words, the partnership is our unit, while the involved partners are referred to as subunits, as seen below:

Unit of analysis: The partnership/education project Subunits of analysis:

Southern partners

 RECOMPAS (local CBO)

 Education Secretary of Tumaco (SET)

 National Ministry of Education (MoE)

 University of Nariño (Udenar)

 National University Open and at Distance (UNAD)

 University of Cauca (Unicauca)

 Education Institutions (EIs)

 2 Colombian NGOs

Northern partners

 Norwegian Refugee Council (NRC)

 Save the Children in Colombia (SCiC)

 Canadian International Development Agency (CIDA – donor)

Table I: The unit and subunits in the embedded single-case design

Although there are many subunits, the focus for the investigation is primarily on the relation between NRC, SC and the local CBO, la corporación Red de Consejos Comunitarios del Pacífico Sur, the Network of Community Councils of the South Pacific Corporation (RECOMPAS), and to a lesser degree includes SET, and MoE (the local and national public education authorities).

(30)

14

According to Yin, a single-case design can be used when the case is either: (a) a critical test of existing theory, (b) a rare or unique circumstance, or (c) a representative or typical case, or where the case serves a (d) revelatory or € longitudinal purpose (2009, p. 52). As for our case, we want to test an existing theory with a representative case, so we have several of the conditions to complete a single-case study. If we would have more time and resources available, a multiple-case design would be preferable, linking this to the previous discussion on the problem of generalisation and theory construction originated from just one case. As we discussed, the single-case strategy puts limitations on the possibility to generalise around a single case study, since the conclusions originating from this type of studies are much stronger if they can be compared to other studies, although these limitations are not categorical (George & Bennett, 2005, p. 220). We could have run several (multiple) case studies on the same subject of North/South partnerships for education, but in different geographical locations in the country and with different partners. This would be a procedure to replicate a case study, compared to the replication of an experiment. This procedure could strengthen or weaken our results based on the possible outcomes.

2.2.4 The Logic of Linking the Data to the Propositions and Criteria for Interpreting the Findings

The fourth step in Yin’s research design relates to how data should be collected in order to analyse it, which in turn will lay the groundwork for the conclusions and results. This is also linked with the fifth step, were we need to arrange some criteria for how to interpret the results.

Operationalisation

As a way of concretising the research questions for the study, we demarcate the area in focus of the study into specific categories which are easier to measure in our data collection phase (Johannessen et al., 2010, p. 63). This is also known as the operationalisation of a study. The mentioned categories are known as indicators for the research, which were briefly presented in the introductory chapter.

(31)

15 Research indicators

It is necessary to elaborate different indicators that will contribute to sustain or reject the research questions, giving us the needed answers. Indicators typically form simplified representations of a complex phenomenon, and are easier to respond to than the research questions themselves. According to Johannessen et al (2010, p. 64), it is unmanageable to investigate “everything”. Therefore, it is necessary to construct indicators that guide the research process. Indicators are representative to the phenomena being investigated.

In the planning stage for creating the indicators, we have to keep in mind that the two NDNGOs are well-experienced in that their development projects are being evaluated by the donors and external entities. In this case, as part of the sources used for the investigation, external evaluations have been used to better understand how the projects are functioning, from a donor’s point of view. Therefore, it has never been the goal of this study to assess the project in a similar way as done previously by other controlling entities. On the other side, the local CBO has lesser experience in project management, and it is possible that they have a different culture of project planning and the evaluation of project achievement. Thus, methodologically, the indicators help to measure the level of affirmation for the different subjects. For instance one measures how the partners think of the idea of partnership, and how they deal with this in practice. If the partners agree on a common understanding of partnership, it will contribute to confirm the actual co-operation between them.

The chosen indicators are selected from NGO partnership theory presented in the next chapter. As the reader knows, this thesis seeks to outline the relations between international and national organisations working with education in Colombia. Since we have stated that there can be unbalanced relations between southern and northern organisations, we outline in more detail our indicators which will guide the design of the case study:

 Understanding of partnership - How does each partner understand the established partnership?

 Equality in decision making - What are the comments among the partners on the issue of equality in terms of decision making and planning

 Local empowerment - According to the partners, to what degree is the partnership leading to local empowerment?

(32)

16

 Co-ordination - Who is responsible for the implementation of necessary activities related to the project? How is the communication between the partners?

 Communication and transparency – How transparent is the partnership in terms of information exchange and how (often) does communication occur?

 Expectations and common goals – to which degree can it be said that the partners share a mutual understanding of project outcomes?

 Challenges – according to the partners, what are the main challenges in the partnership?

 Sustainability – what can be said about the sustainability of the North/South co- operation, and what are the opinions of the partners about continued co-operation, as well as project sustainability?

The above mentioned indicators are used to determine our results. For this, we have to give them certain values. To elaborate on this, we will show some examples of how we measure our indicators. For example, considering the first indicator partnership, the researcher is looking for how each partner understands the concept. If the partners agree on the understanding of partnership, the indicator contributes to strengthen the belief that the relationship is working. For the next indicator equality in decision making, we look for equality between the partners in terms of this issue. Thus, if both southern and northern partners interfere on an equal basis, the partnership model is further strengthened. If the partners have frequent communication, the communication is defined as good. A final example is the indicator local empowerment. For many development projects, this issue is considered central for a well-functioning partnership. If the data shows that the partnership has increased knowledge, technical management and other human resources with the local partners such as the local CBO and education authorities, the indicator will sustain that the NNGDOs are transferring specific knowledge that the local partners need in order to continue the project in a later stage without the presence of the former entities.

Lastly, the indicators help us to establish the level and type of partnerships that exist between the northern and southern organisations, following the theory of Alan Fowler. The partnership concept is discussed in further detail in the next chapter.

(33)

17 Rival explanations

Apart from the indicators, Yin (2009, p. 34) states that an important strategy is to elaborate rival explanations for the findings, something that we do prior to the data collection. As was presented in the introduction chapter, this forms an alternative to show statistical significance using p-values, which would better suit a quantitative study. This will augment the study’s validity and reliability. Therefore, we think of the other possibilities that can affect how the relationship is working in practice. For instance, the context in which each partner is situated can affect how the partnership is perceived. It would be natural to assume that the partners have different perspectives on the partnership according to their background and prior experiences. The relationships with public entities can also be stronger than one assumes. It is also possible that the NGOs have high levels of freedom in terms of project design and implementation, without being restrained from the donor. It is also possible that the NGOs have long traditions of co-operation with other NGO and public partners.

2.3 Case study

The case study method has more or less been incorporated as a complement to other existing methods, but applicable literature is now readily available on the matter, and although there still are some discussions on its validity and reliability, it has gained acceptance as an alternative way to elaborate a research project. In this part, the work of Robert Yin, Case Study Research (2009), is mainly used as a source for our research design. Basically, the case study method is understood as: “an empirical inquiry that (a) investigates a contemporary phenomenon in depth and within its real-life context, especially when (b) the boundaries between phenomenon and context are not clearly evident” (Yin, 2009, p. 18). Apart from this, it also serves to structure the whole process from the development of theoretical propositions to guide data collection and analysis. More than simply guiding specific data collection techniques, case study is an all-compassing model to structure a complete research project.

2.3.1 Considerations on the use of the method

Although the case study method receives much positive appraisal, it also receives criticism.

One central concern refers to a lack of rigour, stating that the method is not used in a systematic way, or that the findings may be biased to correlate with desired outcomes. It is argued that this lack of rigour can be avoided through the use of other methods, that possibly

(34)

18

are easier to use due to easier access to detailed texts helping to structure the research from A to Z, something that case study is lacking, considering that there are fewer writings available on the subject (Yin, 2009, p. 14).

Another common critique is that case studies provide little basis for scientific generalisation, if one considers that one has to do with a single case or a very limited number of cases. This critique has been countered with another question: “How can you generalise findings from a single experiment?” This statement highlights that scientific facts rarely are based on just one single experiment, but rather on several of them. Rather, it is more accurate to say that case studies and experiments both serve to generalise on theoretical propositions, and not on populations or universes (Yin, 2009, p. 15).

A third criticism towards the case study method is directed to the process as being too long, which in turn leads to high amounts of unreadable documents. According to Yin, this aspect can have some relevance to former case studies, but he also asserts that this has to do with the misinterpretation of the case study method as being an ethnographic one, an approach that usually includes long periods of time working in the “field”. Rather, Yin claims that it is possible to do a valid and high-quality case study from ones home office using the Internet or the telephone, all depending on the scale of the investigation. In many ways, instead of assessing the method as an alternative approach, it should rather be seen as a complementing methodology to more traditional ones, such as the conduction of experiments. Anyhow, a successful implementation of a case study implicates a well-structured design (Yin, 2009, pp.

15-16).

2.3.2 Confining the case

Our case depicts a partnership between units that mainly works on a local level in the Tumaco municipality in the Nariño department of Colombia, although the partnership comprises education projects in several areas of both the Nariño and Cauca departments, but the study focuses mainly on the Tumaco, since it gives a good representation of the functionality of the partnership.

The partnership refers to two development projects, in which most of the former mentioned subunits are active. The first project was launched in 2008. Thus, in a time frame, the study

(35)

19 focuses on a partnership that has existed since 2008, and includes the initial start-up phase of the second project, which took place in 2013.

2.4 Field work

Compared to a quantitative method, when collecting qualitative data the researcher’s role is more prominent in the data collection, meaning that the process of the data collection is more difficult compared to the handling of quantitative data, referring to the issue of the instrument used being less elaborated and less established on beforehand. On the other side, when collecting quantitative data, the measurement instrument is the most important link in the design. Such instruments have to be thoroughly elaborated, structured and tested in advance to assure their effectiveness. Generally, the implication of the researcher is more important during the collection of qualitative data compared to quantitative data (Grønmo, 1998, pp. 90- 91).

2.4.1 Types of Data Sources

Concerning secondary sources, we have mainly used documents written by the two northern NGOs, as well as some UN and official Colombian sources. Mainly internal non-published and published documentation was collected from several partners, particularly from NRC and SCiC. In lesser degree documentation was collected from MoE, SET, RECOMPAS and the co-operating universities. We have also included sources that stem from both external and internal evaluations of the development projects that both NRC and SC are running in Colombia.

Johannessen et al (2010, p. 86) state that it is an advantage to combine multiple data sources when conducting a case study, where the goal is to obtain as much data as possible on the chosen subject. The use of multiple sources also constructs validity for the study. As for primary sources, mainly interviews were used, including non-formal conversations and focus groups interviews. Participatory and non-participatory observations were also used for data collection. The initial plan was to include all the partners involved in the partnership, though some of them are more relevant than others due to their proximity and their involvement in the partnership for education. Therefore, the same subunits that comprise the partnership as

(36)

20

listed earlier also serve as primary sources for the data collection. As for IEs, four local schools serve as sources for interviews and observation.

Apart from consulting the regional partners in South-west Colombia, interviews were also conducted with NRC staff in Oslo, Norway, NRC staff in Bogota, NMFA in Bogota, and staff at a Colombian NGO working with education in another department of the country. During the planning and conduction of the field work, the researcher chose to focus primordially on the relations between NRC, SCiC, RECOMPAS, the EIs, SET and the universities, since their relation reflects the main North/South divide and therefore best represents the topic of the study, according to the investigator. Finally, apart from the sources being categorised as less relevant, due to constraints with time and economical resources, no interviews were conducted with CIDA or MoE, which are situated in Bogota.

2.4.2 Method for Data Collection

In the planning phase of the study, several documents serving as a protocol for the study were designed to help the investigator with keeping an overview, and serve to maintain and structure the study. The documents also serve to connect the different stages of the project, expected to increase the validity of the investigation. In these documents, we wrote down the research questions, rival explanations and our sources, as well as a plan for how we are to collect the necessary data. Additionally, a database with raw-data was created to keep track of the primary sources. All the former measures strengthen the validity and reliability, since it makes our case study more transparent and makes it possible to trace the different phases of the project, something that also increases the possibility to monitor the theory afterwards (Yin, 2009, p. 45). It is preferred to have several sources stating the same with similar outcomes. Ideally, the investigator wants sources with contrary tendency; if two sources with opposing point of departure assert the same meanings, the confidence towards the statement is strengthened (Kjeldstadli, 1999, p. 178).

Interviews

The indicators and study objects were applied to design the interview guides. Moreover, the subjects also have relevance for the supplementary data collection methods. The advantage of using interviews is that they are insightful and goal-oriented; they focus directly on the subjects of the case study. Weaknesses regarding this strategy are that the answers can be of

(37)

21 bad quality, especially when the questions are poorly formulated. The possibility of information errors or even direct lies in the answers, as well as the risk that respondents may reply what he or she considers to be what the interviewer wants to hear, are other disadvantages of interviews for data collection according to Yin (2009, p. 102). Moreover, the enactment of the interview itself also needs to be addressed in an ethical way; for instance we need to assure that we avoid the use of leading questions, while at the same time evade asking questions that can be understood as threatening, since good answers depend on a good relation with the interviewee.

Considerations of the use of interviews

Furthermore, it is necessary to request consent from the respondents, as well as to examine Colombian regulations for data collection and storage. As for this particular issue, a formal consent letter was elaborated in Spanish and used in most of the interviews, except from a few made with NRC and SCiC staff, where it was considered not to be necessary when asked. The informants were also guaranteed full discretion in terms of that their names would not appear in the final work. The applied letter of consent is found under attachments. The informants were also informed about the project and how their data would be used. The investigator brought a letter from his university faculty department to gain confidence. This was translated into Spanish and worked as a formal confirmation of the fieldwork. Some of the interviewees showed an interest for the final results of the investigation. As for the issue of data collection and permits, the investigator contacted the Colombian embassy in Oslo with the inquiry. The reply was that it was not necessary to formally request permission to complete a master’s study in Colombia with the data collection method presented.

Interviews are a common strategy for the collection of qualitative data. It was chosen to use semi-structured interviews since this technique makes it possible to highlight the complex social phenomenon that is the subject of the study. If we were to use a standardised design, we would leave out flexibility, which could lead to important information not being collected (Johannessen et al., 2010, p. 138). Several interview guides were designed; one for RECOMPAS, school staff and universities, another for NRC and SC, and a third for SET.

These guides are also among the attachments. In designing the guides, it was important to avoid mixing the two types of questions: (1) The research questions, and (2) the questions directed towards the interviewee, since they are composed differently, and last but not least

(38)

22

we cannot expect that the interviewee is able to respond directly to the research questions (Johannessen et al., 2010, p. 136). The guides are based on a template found in Johannessen et al (2010, pp. 141-142), which categorises the interview in the following sections: (1) Introduction, (2) Fact questions, (3). Introductory questions, (4) transitional questions, (5) key questions, and (6) closing questions.

In addition to interviews, direct observation, participant-observation, and informal conversations were also used. The advantage of these approaches is that they cover events in real time in their real surroundings, as well as providing contextual information on the case.

However, these strategies are time consuming, and can also lead to selectivity, meaning that the researcher fails to observe what really should be observed. Moreover, the investigator also risks to bias the events with his or her presence (Yin, 2009, p. 102).

Selection of sources

Since we deal with several units, where every one of them consists of many people, it is essential to make a selection, also known as sampling for the data collection.

Methodologically, sampling should be random to avoid selecting only those expected to confirm our hypothesis. At the same time, one needs to assure that the selected participants are representable for the case. Therefore, it makes sense to collect data from the people who in different ways have been active in the partnership for education, for example the NGO and CBO staff working directly with the education project. In a certain way, considering the difficulty for the investigator to gain access to relevant sources, the snowball-method was used, meaning that the researcher asked NGO staff involved in the project to help him get in touch with relevant people. A certain problem with this strategy is that NGO staff can select informants that represent certain preferred opinions, leaving out others. Still, this is not considered to be a big risk, also since there are only few people involved in the project.

When it comes to interviews, there is no limitation on how many the researcher may need. As a general rule, it is said that 10 to 15 are common, or one completes the task when the data is saturated and the investigator no longer receives new information (Johannessen et al., 2010, p.

104). In total 33 persons gave their feedback, where 23 persons were recorded in 13 interviews. Five of the interviews were with two persons or more. The interviews usually lasted for about an hour. The other persons were consulted through informal conversations.

Partnership for Education? A Case Study on Tumaco, Colombia on a North-South Co-operation for Education in Conflict Areas

Partnership for Education?

A Case Study on Tumaco, Colombia on a North-South Co-operation for Education in

Conflict Areas

Tobias Immanuël Alexander Wilbers

Master’s Thesis in Latin American Studies

Department of Literature, Area Studies and European Languages

Faculty of Humanities

Supervisor: Jemima Garcia-Godos UNIVERSITY OF OSLO

Partnership for Education?

A Case Study on Tumaco, Colombia on a North-South Co-operation for Education in

Conflict Areas

By: Tobias Immanuël Alexander Wilbers

Latin American Area Studies

Department of Literature, Area Studies and European Languages

Faculty of Humanities

Supervisor: Jemima Garcia-Godos

UNIVERSITY OF OSLO Turned in May 15th.

2015

Abstract

Resumen

Preface

Table of Contents

List of Abbreviations

1 INTRODUCTION

1.1 Development in the South

1.2 Education for Development

1.2.1 Why is the Subject Important?

1.3 Confining the Case

1.3.1 Research Topic

1.3.2 Hypothesis Setting the Groundwork for the Research

1.3.3 The Research Questions

1.3.4 Justification of Research Site

1.3.5 The Structure of the Thesis

2 METHODOLOGICAL FRAMEWORK

2.1.1 Research Design

2.1.2 Validity and Reliability

2.2 Yin’s Five Point Research Design

2.2.1 Research Questions

2.2.2 The Study’s Propositions

2.2.3 Units of Analysis

2.2.4 The Logic of Linking the Data to the Propositions and Criteria for Interpreting the Findings

2.3 Case study

2.3.1 Considerations on the use of the method

2.3.2 Confining the case

2.4 Field work

2.4.1 Types of Data Sources

2.4.2 Method for Data Collection